Effective congestion control algorithms (CCAs) are crucial for the smooth operation of Internet communication infrastructure. CCAs adjust transmission rates based on congestion signals, optimizing resource utilization and user experience. However, existing studies, both rule-based and learning-based CCAs, often struggle with generalization and underperform when deployed in real-world environments. When applied to unseen network conditions, hand-crafted schemes or pre-trained models may experience significant performance degradation. To address this challenge, we propose MetaCon, a novel adaptive Internet congestion control approach based on meta-reinforcement learning. MetaCon leverages knowledge learned from prior scenarios to quickly adapt to new environments. Experimental results show that MetaCon outperforms existing algorithms by exhibiting superior generalization and achieving better transmission performance across a wide variety of network conditions.

Congestion control (CC) is a cornerstone of network communication systems, facilitating high throughput, low latency, and efficient resource utilization. Despite decades of research, existing CC algorithms (CCAs), whether rule-based or learning-based, still fall short of delivering satisfactory performance across diverse network conditions and application demands. Rule-based CCAs inherently suffer from limited adaptability, as their handcrafted control rules are typically designed for specific scenarios. Recent popular learning-based CCAs exhibit strong potential in high performance and adaptability, but often struggle with fairness and fail to generalize well to unseen environments. To address these limitations, we propose Pisces, a novel learning-based congestion control scheme based on multi-agent meta-reinforcement learning (MAMRL). Pisces formulates the CC task under a novel MAMRL-based control framework to optimize the control policy for high performance, strong adaptability, and inter-flow fairness across various network scenarios. To improve policy generalization, Pisces leverages meta-knowledge learned from prior tasks to enable rapid adaptation to new environments. Extensive experiments demonstrate that Pisces delivers consistent strong performance, excellent adaptability, and fairness across a wide variety of both seen and unseen scenarios, outperforming or matching state-of-the-art CCAs.

This paper aims to address the limitations of the conventional modular multilevel converter (MMC) control strategy in flexible middle-voltage direct current (MVDC) transmission technology with regard to voltage utilization and power transfer. In this paper, a novel control strategy is proposed for the MMC topology in DC transmission systems. This strategy is based on the use of a carrier phase-shifted sinusoidal wave modulation (CPS-SWM) system. The efficacy of this control strategy is evaluated through the implementation of simulation and analytical methods, in terms of technical feasibility, economic benefits, and system performance enhancement.

Patent standardization plays an important role in bridging innovation with industrial implementation.Most existing studies in the field of patent standardization focus on predicting whole patent standardization.They overlook partial standardization, where only certain technical concepts within a patent are adopted by standards.The main challenges in conceptlevel standardization include:1.How to effectively extract technical concepts from patent documents.2.How to accurately predict the standardization timeline of these concepts.However, existing methods still struggle to accurately extract technical concepts from patent texts and lack the capability to model their standardization timelines.To address these challenges, we propose a Patent Concept Standardization Prediction Model.This Model extracts technical concepts from patent documents and predicts their standardization distribution over time.We first extract standardized technical concepts from the standards associated with the target patents and use them as predefined categories for patent concept classification.Next, we calculate the semantic similarity between patent paragraphs and these concepts, assigning each paragraph to its most relevant category.Based on the aligned pairs, we train a multi-class classification model to enable large-scale concept-level normalization across the patent corpus.After extracting patent concepts,we use neural network to estimate the temporal distribution of standardization for each concept based on its semantic embedding.Experimental results show that our framework outperforms baselines across both patent-to-concept classification and concept-level standardization time distribution prediction.

Efficient spectrum resource allocation is essential for vehicular networks to ensure seamless connectivity in dynamic environments. However, conventional mathematical and deep reinforcement learning (DRL) methods suffer from scalability challenges and privacy concerns due to centralized training requirements. In addition, the high mobility of vehicles leads to frequent topology changes and switching, making static spectrum allocation schemes ineffective. To address these challenges, this paper investigates a joint federated learning (FL) and proximal policy optimization (PPO) framework (FL-PPO) for spectrum resource allocation in vehicular networks. In our proposed framework, each vehicle independently trains a local DRL model using PPO. At the same time, the base station implements a federated average (FedAvg) algorithm to aggregate models, thereby enhancing global learning while preserving data privacy. By combining FL with DRL, the proposed method improves scalability, adaptability, and safety in highly dynamic vehicular environments. Simulation results demonstrate that FL-PPO outperforms centralized DRL and heuristic baselines, achieving higher spectrum capability and better packet delivery, making it a promising resource allocation solution for vehicular networks.

Information-Centric Networking (ICN) is a promising future network architecture that shifts the host-based network paradigm to a content-oriented one. Over the past decade, numerous ICN congestion control (CC) schemes have been proposed, tailored to address congestion issues based on ICN’s transmission characteristics. However, several key challenges still need to be addressed. One critical issue is that most existing CC studies for ICN do not consider the diverse Quality of Service (QoS) requirements of modern network applications. This limitation hinders their applicability across various applications with different network performance preferences. Another ongoing challenge lies in improving transmission performance, particularly considering how to appropriately coordinate congestion control participants to enhance content retrieval efficiency and ensure reasonable resource allocation, especially in multipath scenarios. To tackle these challenges, we propose QSCCP, a QoS-aware congestion control protocol built upon NDN (Named Data Networking), a well-known ICN architecture. In QSCCP, diverse QoS preferences of various traffic are supported within a collaborative congestion control framework. A novel multi-level, class-based scheduling and forwarding mechanism is designed to ensure varied and fine-grained QoS guarantees. A distributed congestion notification and precise feedback mechanism is also provided, which efficiently collaborates with an adaptive multipath forwarding strategy and consumer rate adjustment to rationally allocate network resources and improve transmission efficiency, particularly in multipath scenarios. Extensive experimental results demonstrate that QSCCP satisfies diverse QoS requirements while achieving outstanding transmission performance. It outperforms existing schemes in throughput, fairness, delay, and packet loss, with a rapid convergence rate and excellent stability.

This paper explores an innovative approach to enhance the resilience and security of beyond 5G (B5G) networks through the implementation of cross-bandwidth part (C-BWP) frequency hopping at mini-slot granularity. Utilizing dynamic channel estimation, the proposed system assigns resource blocks (RBs) to user equipment (UEs) of varying priorities, mitigating the impact of jamming in hostile radio environments. We introduce strategic C-BWP frequency hopping for high-priority UEs, optimizing the use of unaffected RBs. This method is shown to effectively counter various types of jamming, ensuring robust and secure communication in both current and future cellular networks. Through rigorous simulation, we demonstrate that intra-slot frequency hopping offers superior resilience by adapting quickly to dynamic channel conditions, significantly enhancing the performance and security of the communications system.

This research proposes a dynamic resource allocation method for vehicle-to-everything (V2X) communications in the sixth generation (6G) cellular networks. Cellular V2X (C-V2X) communications empower advanced applications but at the same time bring unprecedented challenges in how to fully utilize the limited physical-layer resources, given the fact that most of the applications require both ultra low latency, high-data rate and high reliability. Resource allocation plays a pivotal role to satisfy such requirements as well as guarantee Quality of Service (QoS). Based on this observation, a novel fuzzy-logic-assisted$Q$learning (FAQ) model is proposed to intelligently and dynamically allocate resources by taking advantage of the centralized allocation mode. The proposed FAQ model reuses the resources to maximize the network throughput while minimizing the interference caused by concurrent transmissions. The fuzzy-logic module expedites the learning and improves the performance of the$Q$-learning. A mathematical model is developed to analyze the network throughput considering the interference. To evaluate the performance, a system model for V2X communications is built for urban areas, where various V2X services are deployed in the network. Simulation results show that the proposed FAQ algorithm can significantly outperform deep reinforcement learning,$Q$-learning and other advanced allocation strategies regarding the convergence speed and the network throughput.

Information-Centric Networking (ICN) has been proposed to provide a scalable and efficient content delivery solution. To support different kinds of applications in ICN, a practical congestion control mechanism should consider various traffic and the corresponding requirements of the quality of service (QoS), which is ignored by most of the existing schemes. In this paper, we propose a differentiated service-based congestion control protocol (DSCCP) for ICN. DSCCP defines the service classes of various ICN traffic and performs the class-based hop-by-hop Interest shaping at intermediate nodes. Meanwhile, it employs an explicit rate feedback mechanism to notify consumers about available network resources. It also provides feedback for the forwarding strategy of the router to adjust the forwarding probability of each interface. Additionally, a punishment tactic is used to tackle misbehaving consumers by limiting the Interest forwarding rate. The simulation results show that DSCCP can ensure the class-based bandwidth allocation among consumers with different service classes, as well as, quickly converge to a high throughput with low latency and low packet loss rate.

Each blockchain’s node needs to store a backup of all blocks, resulting in the whole network needs O(n) storage space, which greatly affects the nodes’ scalability. Parallel proof of vote(PPoV) is a permissioned blockchain algorithm that uses block groups as the basic data structures. This paper proposes a data lightweight scheme for PPoV. In the bock group generation stage, the BLS algorithm is used to realize signature aggregation, and the storage space of block signature and vote signature is reduced from O(n) to O(1) without affecting the performance. In the storage stage, we use erasure code to implement storage partition for block groups and ensure the real-time recoverability of complete data. To speed up reading, the timeline-based model applies different storage strategies for hot and cold data. The experimental results show that the empty block group generated by BLS aggregation signature is smaller and less sensitive to the number of nodes. And the reduction ratio of storage space under a large number of transactions can be similar to the number of normal nodes.

In this paper, we spotlight vehicle-to-everything (V2X) communications in 5G cellular networks. Cellular V2X (C-V2X) communications in 5G enable more advanced services with requirements of ultra-low latency and ultra-high reliability. How to make full use of the limited physical-layer resources is a key determinant to guarantee the quality of service (QoS). Therefore, resource allocation plays an essential role in exchanging information between vehicles, infrastructure, and other devices. In order to intelligently and reasonably allocate resources, a self-adaptive fuzzy logic-based strategy is developed in this paper. To evaluate the network performance for this adaptive strategy, a system model for V2X communications is built for urban areas, and typical safety and non-safety services are deployed in the network. Simulation results reveal that the proposed fuzzy logic-based algorithm can substantially improve resource utilization and satisfy the requirements of V2X services, compared with prior counterparts, which cannot provide guaranteed services due to low resource utilization.

Basic Safety Messaging plays a crucial role to provide road safety in vehicular ad-hoc networks (VANETs). To avoid potential accidents, vehicles periodically broadcast safety information to neighboring vehicles. However, due to transmission collisions, fading channels and other factors, vehicular networks usually suffer a low packet delivery ratio (PDR) and a large delay, which are intolerant of many safety applications. To tackle these issues, this paper proposes a hybrid medium access control (MAC) protocol for basic safety message (BSM) dissemination based on the framework of Dedicated Short-Range Communication (DSRC). Its partially centralized and partially distributed characteristic not only can effectively suppress the collisions, but keep compatibility with IEEE 802.11p. In addition, the integration of Physical-Layer Network Coding (PNC) and Random Linear Network Coding (RLNC) further strengthens the reliability and efficiency for BSM dissemination. Both the theoretical analysis and comprehensive simulations indicate that, compared with existing schemes, the proposed protocol can significantly improve the PDR by a range of 20% to 300%. Meanwhile, in terms of normalized throughput, it increases by varying percent between 20% and 160% in different scenarios.

Basic Safety Messaging (BSM) is a crucial application to provide road safety in vehicular ad-hoc networks (VANET). This paper proposes an efficient and reliable MAC protocol for BSM packets dissemination based on the framework of Dedicated Short-Range Communication (DSRC). It perfectly integrates Physical-Layer Network Coding (PNC) and Random Linear Network Coding (RLNC) in both roadside unit (RSU) and onboard unit (OBU) nodes. Comprehensive simulation shows that compared with the existing schemes for BSM dissemination, the proposed protocol achieves both high flexibility and excellent performance in packet delivery ratio (PDR).